지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학술대회자료
- 저자정보
- 저널정보
- 한국정보통신학회 INTERNATIONAL CONFERENCE ON FUTURE INFORMATION & COMMUNICATION ENGINEERING 2014 INTERNATIONAL CONFERENCE ON FUTURE INFORMATION & COMMUNICATION ENGINEERING Vo.6 No.1
- 발행연도
- 2014.6
- 수록면
- 205 - 208 (4page)
이용수
초록· 키워드
오류제보하기
e-Learning service based on science gateway, supporting multiple users’ public request for simulation jobs in diverse sciences such as computational fluid dynamics, chemistry, physics, and electronics, requires extremely high performance and scalability to store and manipulate a set of scientific meta-data in distributed file systems. Although handling a massive amount of files is inevitable, several file systems previously reported based on the notion of a hierarchical directory structure, show their lack of data manageability and efficient searchability, raising the strong needs for alternatives to handle large-scale file systems. In this work, we propose a data management framework particularly efficient for High Performance Computer (HPC) based e-Learning service using distributed file systems. The suggested framework provides fast meta-data search functionality through a POSIX-compatible file system that can be easily integrated into any HPC legacy codes. Real-time file search is supported using a distributed file system-based index striping scheme to enhance the performance of dynamic index management.
목차
Abstract
I. INTRODUCTION
II. DATA MANAGEMENT FRAMEWORK DESIGN
III. DATA MANAGEMENT OPERATIONS
IV. CONCLUSIONS AND FUTURE WORK
REFERENCES
참고문헌 (0)
참고문헌 신청이 논문의 저자 정보
최근 본 자료
댓글(0)
0
UCI(KEPA) : I410-ECN-0101-2018-004-000962440